{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:OSFT67S7N7ZDWKVW255OHPVINP","short_pith_number":"pith:OSFT67S7","schema_version":"1.0","canonical_sha256":"748b3f7e5f6ff23b2ab6d77ae3bea86bcb4db8046ff5146392c5d18ad3f08807","source":{"kind":"arxiv","id":"2607.15732","version":1},"attestation_state":"computed","paper":{"title":"IoUPD: IoU-Aware Privileged Distillation for Visual Grounding with Multimodal Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Dongming Zhang, Hao Wu, Jian Xue, Ke Lu, Xiuyuan Zhu, Zijin Du","submitted_at":"2026-07-17T08:13:01Z","abstract_excerpt":"Visual grounding with multimodal large language models is commonly formulated as autoregressive coordinate generation, where a model outputs bounding-box coordinates as text given an image and a referring-expression prompt. While this interface is simple and compatible with instruction following, it introduces a mismatch between training and evaluation: training optimizes token-level likelihood over coordinate strings, whereas grounding quality is measured by geometric overlap. We propose IoUPD, an IoU-aware privileged distillation method for coordinate-generating multimodal large language mod"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2607.15732","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-07-17T08:13:01Z","cross_cats_sorted":[],"title_canon_sha256":"788485f9cc744c33ab58e08afa5b566e4002bcf855bd741adb2aae6b72b6c452","abstract_canon_sha256":"6ab192fd0c1f850811fcf4df3c73ca520f1a1c959df33b7e8cd54bd8851e4300"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-20T01:19:06.820022Z","signature_b64":"zqTGJ+feVrWMOvY7G6qSn/BWd4ukXB7TF/iFItZOb9FV6z01c27nYTtOfgFyUV5+CM6aNei1gxnzX2JPRGNxBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"748b3f7e5f6ff23b2ab6d77ae3bea86bcb4db8046ff5146392c5d18ad3f08807","last_reissued_at":"2026-07-20T01:19:06.819109Z","signature_status":"signed_v1","first_computed_at":"2026-07-20T01:19:06.819109Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"IoUPD: IoU-Aware Privileged Distillation for Visual Grounding with Multimodal Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Dongming Zhang, Hao Wu, Jian Xue, Ke Lu, Xiuyuan Zhu, Zijin Du","submitted_at":"2026-07-17T08:13:01Z","abstract_excerpt":"Visual grounding with multimodal large language models is commonly formulated as autoregressive coordinate generation, where a model outputs bounding-box coordinates as text given an image and a referring-expression prompt. While this interface is simple and compatible with instruction following, it introduces a mismatch between training and evaluation: training optimizes token-level likelihood over coordinate strings, whereas grounding quality is measured by geometric overlap. We propose IoUPD, an IoU-aware privileged distillation method for coordinate-generating multimodal large language mod"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.15732","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2607.15732/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2607.15732","created_at":"2026-07-20T01:19:06.819586+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.15732v1","created_at":"2026-07-20T01:19:06.819586+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.15732","created_at":"2026-07-20T01:19:06.819586+00:00"},{"alias_kind":"pith_short_12","alias_value":"OSFT67S7N7ZD","created_at":"2026-07-20T01:19:06.819586+00:00"},{"alias_kind":"pith_short_16","alias_value":"OSFT67S7N7ZDWKVW","created_at":"2026-07-20T01:19:06.819586+00:00"},{"alias_kind":"pith_short_8","alias_value":"OSFT67S7","created_at":"2026-07-20T01:19:06.819586+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OSFT67S7N7ZDWKVW255OHPVINP","json":"https://pith.science/pith/OSFT67S7N7ZDWKVW255OHPVINP.json","graph_json":"https://pith.science/api/pith-number/OSFT67S7N7ZDWKVW255OHPVINP/graph.json","events_json":"https://pith.science/api/pith-number/OSFT67S7N7ZDWKVW255OHPVINP/events.json","paper":"https://pith.science/paper/OSFT67S7"},"agent_actions":{"view_html":"https://pith.science/pith/OSFT67S7N7ZDWKVW255OHPVINP","download_json":"https://pith.science/pith/OSFT67S7N7ZDWKVW255OHPVINP.json","view_paper":"https://pith.science/paper/OSFT67S7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.15732&json=true","fetch_graph":"https://pith.science/api/pith-number/OSFT67S7N7ZDWKVW255OHPVINP/graph.json","fetch_events":"https://pith.science/api/pith-number/OSFT67S7N7ZDWKVW255OHPVINP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OSFT67S7N7ZDWKVW255OHPVINP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OSFT67S7N7ZDWKVW255OHPVINP/action/storage_attestation","attest_author":"https://pith.science/pith/OSFT67S7N7ZDWKVW255OHPVINP/action/author_attestation","sign_citation":"https://pith.science/pith/OSFT67S7N7ZDWKVW255OHPVINP/action/citation_signature","submit_replication":"https://pith.science/pith/OSFT67S7N7ZDWKVW255OHPVINP/action/replication_record"}},"created_at":"2026-07-20T01:19:06.819586+00:00","updated_at":"2026-07-20T01:19:06.819586+00:00"}